Saturday, November 30, 2013

Cryptocurrencies, Mobile Payment Systems, and The Third World


M-Pesa is an electronic payment system popular in the third world.  It lets people send and receive payments with their cell phone.

It is a roaring success because A: mobile phones are a roaring success and B: most banks, payment products, and financial networks are a roaring failure in the developing world.  M-Pesa succeeded where banks failed.  Weird.

One might almost get the idea that the third world didn't trust banks.  Another theory might be that the poor in the third world were just too poor to nickle and dime to death with fees like the poor in the US, because they remained unbanked for a long time.

However, the poor in the third world were wealthy enough to buy some cell phone time, and began bartering that time in the early 21st century, effectively inventing their own electronic currency.  Smarties.

Seeing an opportunity, mobile phone provider Safaricom invented M-pesa, a mechanism for value exchange on their network.

Safaricom did some neat things with M-pesa, including innovating in microfinance, and involving entrepreneurs as M-pesa Agents - like little one-person bank branches.

Before M-pesa, Kenya was probably typical of African countries - more than 80% of the population was unbanked.(2)  None of those people used credit cards.

Now over 70% of households use M-Pesa, and many are getting traditional banking services through the M-Pesa system.  They still don't use credit cards.  Smarties.

All of their small electronic transactions are in M-pesa.  In June 2010, amazingly only three years after its intro, M-PESA transactions amounted to about 70% of the volume of electronic transactions in Kenya by volume, but only 2.3% by value. (1)

So it appears the third world is getting a good system for daily transactions that in many ways leapfrogs our dated credit-card-based system in functionality.

There are numerous advantages to their system.  Two obvious ones are that it's more convenient for lots of transaction types - you always have a phone and never have a bank or ATM when you need one; and the fees are more reasonable than most of the previously established systems for electronic funds transfer(3)(4)(5).

There are less obvious advantages, and much more interesting ones.  Take the case of the introduction of M-Pesa into Afghanistan:

"In 2009, the Afghan National Police began a test to pay salaries through mobile telephones, rather than in cash. It immediately found that at least 10% of its payments had been going to ghost policemen who didn’t exist; middlemen in the police hierarchy were pocketing the difference. Salaries for Afghan police and soldiers are calculated to be competitive with Taliban salaries, but beat cops and deployed soldiers had been receiving only a fraction of the amount paid by US taxpayers because of corruption in the payment system. Most Afghan cops assumed that they had been given a significant raise, when, in fact, they simply received their full pay for the first time--over the phone."(6)

Interestingly, the spread of M-pesa from Kenya to other third world countries appears to have introduced a reliable ledger and payment delivery system to organizations who did not have one.  Now they have one.  They pay their workers directly.  So do most employers who use M-pesa.  Pretty much everywhere it's used.

This makes me think it's just the tip of the iceberg for Safaricom.

I also think this level of prevalence of mobile payments creates a different mindset in it's users.

What happens when you want to pay one of your workers in the US?  You do it the old fashioned way.  The process gets touched by HR, Accounting, Legal, and probably by a payment processing corporation like ADP before the federal, state, and local agencies even look at it.  Eventually it ends up as a paper check in your employees hands or in their bank, a little lighter than you or your employee might have liked.

An entrepreneur in the US would expect to go about improving that system by incrementally relieving a pain point within it.  Entrepreneurs growing up with M-Pesa see this as a feature of a mobile app - they would expect the addition of a "this is for payroll" checkbox during a transaction would be enough to add a note to the transaction log - further accounting to be handled by any number of third party applications, or by themselves, or by Safaricom itself.

While the third world advances, the first spins it's wheels.  Numerous abortive attempts have been made to build similar or superior infrastructure to M-pesa (5).  Established players in banking and payment systems are scrambling to compete (see PingIt, Square, etc), but all are clunkier than M-pesa, and nothing has taken hold.

Part of the reason is just momentum, once dipped in the warm water of broadband and credit/debit cards, we see no reason to stop buying on Amazon with our laptop, or with our credit card at the grocery store.  We have a familiar, working solution in broadband and credit cards, although they cost us dearly.  As a group, we don't care all that much that merchants and expats get screwed by fees, and we just can't measure the impact to ourselves at the register.  Anyway, nuff' said.  We're missing another wave.  We'll get there, but we'll be toting the baggage of our old industries across the finish line in last place.  Back to the people who are doing it right.

I see a couple things happening: Mobile payment systems rapidly spreading across the third world, and, longer term, mobile payments system users getting heavily into cryptocurrencies.

Not being based in the third world, it's not a safe bet for me to predict anything about it, but I feel pretty good about those two bets, as they are already happening.  Kipochi is a bitcoin wallet that already integrates an M-Pesa-bitcoin gateway.  (9)  I'm sure there are others in the works, but I haven't looked.  M-Pesa is profitable (13) and fueling competition. (14)(15)

Why would an M-Pesa user be interested in Bitcoin?  Well, have you ever run an online store?  If you have, your anti-fraud software probably made it difficult for the third world to buy from you.  For good reason.  Statistically, you never want to accept a credit card from someone you don't know in the third world.  Those transactions are fraud 90% of the time.  The transaction cost is ridiculous to sell to Nigeria, because you personally have to talk to the person, and somehow work out whether they are honest before you can sell to them.  The lack of security inherent in credit card transactions creates enough friction to price the poor right out of the online transaction world for all but the largest, highest touch transactions (which they are not going to make, so they're out!).

Since Bitcoin solves the third worlds international fraud problem(10), someone in Africa using Kipochi can buy something from any online vendor that accepts Bitcoin[B].  A whole new world is about to open up to them.  Bitcoin will catch on in the third world (or some cryptocurrency will).  It might eat into Safaricoms transaction fees, but not too much.  It only makes sense to use Bitcoin for larger transactions.  Besides, the mobile carriers need something that will push all transactions onto the web eventually.  They are in the business of selling broadband, too.  If all their customers want to create online websites and sell product to one another for Bitcoins, it's a huge win for them.  Who knows, Safaricom might even strike a deal with Amazon to have their M-pesa Agents hold packages for people.

Obviously, Africa has to fix a few other things, but Amazon.com might be in there sooner than you think.

I'm serious.  Bitcoin, or something like it, by taking the fraud concern, friction, and transaction log concerns out of international commerce, fits M-Pesa like a glove, and could help to bring the third world into the first.  Bitcoin, plus rapid innovation in mobile payments, could even put the third world in an enviable position compared to the first with regard to financial technology. 

So who's killin' it in the third world?  Mobile carriers.  Did I just say that?  I know, it just seems nuts to anyone in the US.  The average US consumer knows that mobile carriers and banks will fight to the last breath to keep the status quo and kill innovation.  Innovation is pretty tough to kill...hmmm...excuse me while I grab some popcorn and watch that fight.  Keep fighting innovation mobile carriers and banks!  When the dust settles maybe we can get M-Pesa. [C]

(1) http://www.cgap.org/blog/10-things-you-thought-you-knew-about-m-pesa
(2) http://www.businessweek.com/articles/2013-03-06/what-africa-can-teach-us-about-the-future-of-banking
(3) http://www.safaricom.co.ke/personal/m-pesa/m-pesa-services-tariffs/tariffs
(4) http://www.mit.edu/~tavneet/M-PESA.pdf (page 7)
(5) http://usa.visa.com/download/merchants/Interchange_Rate_Sheets.pdf
(6) http://smallwarsjournal.com/jrnl/art/one-cell-phone-at-a-time-countering-corruption-in-afghanistan and http://en.wikipedia.org/wiki/M-pesa#Afghanistan
(7) http://www.fiercemobileit.com/story/nfc-mobile-payments-flop-says-gartner/2013-06-04
(8) http://www.mit.edu/~tavneet/M-PESA.pdf
(9) http://www.kipochi.com
(10) I'm going to cop out and refer you to this, which you should read and understand: http://bitcoin.org/bitcoin.pdf
(11) http://www.gsma.com/mobilefordevelopment/programmes/mobile-money-for-the-unbanked/mmu-examples/m-pesa
(A) 1 US Dollar is about 85 Kenyan shillings and about 1 milli-bitcoin.  At those ratios M-Pesa limits money out of a phone wallet to about one bitcoin per day.
(13) http://www.gsma.com/mobilefordevelopment/safaricom-m-pesas-h1-fy13-results-a-portrait-of-a-maturing-mobile-money-service
(14) http://techloy.com/2012/12/24/paga-reports-790-user-growth-in-its-first-year-faster-than-m-pesa-in-kenya-and-tanzania/
(15) http://www.businessdailyafrica.com/Corporate-News/Airtel-targets-MPesa-with-Equity-Bank-agency-deal/-/539550/1931908/-/93q174z/-/index.html
(16) Also interesting: http://www.gsma.com/mobilefordevelopment/programmes/mobile-money-for-the-unbanked/mmu-examples/m-pesa
(B) Yeah, I'm assuming that most major vendors will start to accept bitcoin and/or the gateway cost will approach zero.  Right now that's possible with Bitcoin Anything, etc., and it should get cheaper over time.
(C) That seemed like a nice way to end this blog post.  In reality, it's more complicated than that.  M-Pesa is dragging all other payment networks into the third world to compete.  What will happen in the first world is that we will need an equally innovative system to move our entrenched markets about.  I will make the popular contention, (among the other two readers of this blog) that some good old fashioned free market competition is in order to eliminate friction and grease the wheels of innovation.